Spatiotemporal Analysis of LULC Dynamics in Ramsar Wetlands Using GEE: A Multi-temporal Assessment of the Keta Lagoon Complex and Muni-Pomadze Ramsar Site.

Keywords: LULC, GEE, machine learning, coastal wetland, climate change.
Abstract Type: Paper Abstract

Abstract

Coastal marshes and Ramsar sites are critical for biodiversity and ecosystem services but are increasingly vulnerable due to human activities and climate change. This study focuses on developing a model to track land use and cover (LULC) changes over time at the Keta Lagoon Complex and Muni-Pomadze Ramsar Site in Ghana, using Google Earth Engine (GEE). By leveraging satellite imagery and GEE's cloud computing capabilities, this research efficiently processes large datasets from Landsat and Sentinel collections to analyze historical and current LULC trends. Machine learning algorithms enhance classification accuracy, using indices like NDVI and MNDWI to differentiate wetlands, vegetation, and urban areas. In addition, the study integrates climate and hydrological data to explore connections between LULC shifts and environmental factors, shedding light on how natural processes and human impacts contribute to wetland vulnerability. The model provides a robust, adaptable framework for monitoring LULC in Ramsar sites facing similar environmental challenges, producing time-series LULC maps that aid in conservation decision-making. This approach not only supports conservation strategies for Ghana’s wetlands but also sets a precedent for employing GEE in ongoing ecological monitoring. The findings offer conservation organizations valuable data to inform actions, and the model’s adaptability makes it a valuable tool for wetland preservation amid global environmental change.